What causes shoulder impingement, and what makes it flare up

What causes shoulder impingement?

What causes shoulder impingement, in most shoulders, is not one thing you can put your finger on, and the honest version of that answer is more useful than the tidy one. The Dutch Orthopaedic Association guideline, which is the guideline this page leans on most, says the anatomical explanation of a pinched rotator cuff "is not sufficient to cover the pathology", and that a direct relationship between the anatomy, the load on the shoulder and the pain is not always there to be found (Diercks and colleagues, 2014). The 2026 update of that guideline starts from an even plainer admission: there is no universal definition of the condition, so the working group wrote one, and what it settled on describes the symptoms rather than a cause, shoulder complaints that are not traumatic and not rheumatological, with pain mainly when you raise the arm (Lambers Heerspink and colleagues, 2026).

So here is the short version. What the readable evidence supports is a tendon and a bursa in the space under the bony tip of the shoulder that are not coping with what is being asked of them at the moment, in a shoulder whose tendons also change over the years. How much the arm has been doing is part of it. Age is part of it. What the name describes, a bone pinching a tendon as the arm lifts, is the part that has held up worst under testing.

One word about the words, because on this page the name is part of the subject. Your letter may say impingement, subacromial pain or subacromial pain syndrome, which is the term both Dutch guidelines use, or rotator cuff tendinopathy, or rotator cuff related shoulder pain. The shoulder in front of the clinician is much the same one whichever is written down, and the name change is set out with its sources on the main shoulder impingement page, while the tendon itself has its own page at rotator cuff tendinopathy.

Why the name says pinching and the evidence does not

Impingement was a model before it became a diagnosis, and the model is why the word exists at all. The Dutch guideline describes the original concept of an impingement syndrome, caused or made worse by contact between the acromion and the rotator cuff as the arm lifts, and then says that hypothesis cannot be substantiated with better imaging and arthroscopic techniques (Diercks and colleagues, 2014). The same introduction notes where the thinking went instead: more value is now placed on degeneration of the rotator cuff tendons.

The operation the model produced tells the same story from the other end. Shaving bone and soft tissue from under the tip of the shoulder was designed to make room, and the 2026 Dutch update reports that strong evidence has emerged showing no primary role for subacromial decompression surgery in these shoulders (Lambers Heerspink and colleagues, 2026). The placebo-controlled trials behind that position, and what they mean if surgery has been mentioned to you, are on the main shoulder impingement page rather than here.

That is not the same as saying nothing inside a shoulder ever touches anything else. The point is that contact has not been shown to be the thing that produced your pain, and that an explanation which sounds obvious has not survived being tested. Where our sources leave a mechanism unsettled, as they do this one, the page says so and leaves the judgment about your own shoulder with the clinician who can examine it.

What seems to matter: how much, how fast, how high

The prevention chapter of the 2014 guideline is where the readable detail sits. It reports associations between getting this condition and five features of work: repetitive movements of the shoulder or the hand and wrist, work that needs much or prolonged strength from the upper arms, hand and arm vibration where the vibration is high or the exposure long, working with a poor ergonomic shoulder position, and a high psychosocial workload (Diercks and colleagues, 2014). Those are associations measured across groups of workers. They are not a measurement of what caused one person's shoulder pain, and the guideline does not offer them as one.

The 2026 update is blunter about how thin the ground is. Its search for studies on preventive measures found none that met the selection criteria, so its advice to patients rests on tendon recovery principles, biomechanics, clinical reasoning and expert opinion rather than on trials, which the authors say openly (Lambers Heerspink and colleagues, 2026). Their practical line is the one worth keeping: avoid activities that are too long, too heavy or too frequent.

In clinic, that is almost always what the history turns out to contain, and we offer this as practice experience rather than as a published finding. Four weekends of painting a ceiling. A return to overhead pressing after six months away, at the weight that felt normal before. A new job with the arm held out at a bench all day. A baby who now weighs twice what she did and gets lifted out of a cot thirty times a day. The amount matters less than the change, because shoulders tolerate what they are used to. Pressing and lifting are covered on shoulder pain on the bench press, and the desk on neck and back pain from desk work.

Does the shape of the bone cause it?

This is a fair question, because many people have been shown their own X-ray and told their acromion is hooked. The 2014 guideline reports that a correlation has been described between the shape of the acromion, the angled type III shape in particular, and the presence of rotator cuff injuries, and then adds the part that rarely gets passed on: that association is not significant in patients over 50 (Diercks and colleagues, 2014). Its prognosis chapter lists type II or type III shape among the features that go with a worse outcome, on one of its lower tiers of evidence. OrthoInfo notes that a particular X-ray view will sometimes show a small bone spur on the front edge of the acromion.

Read together, that is an association which fades in exactly the age group where tendon findings are most common, not a settled cause. There is also a plain logical problem with blaming the shape: the acromion you have today is broadly the acromion you had last year, when the shoulder was fine, so a fixed shape cannot explain why the pain arrived in March. Whether yours is worth any attention is a question for the clinician looking at your shoulder and your images, and whether images are needed at all is covered on do you need a scan for shoulder impingement.

Does age change the tendon?

Yes, and that is ordinary, not a verdict on how you have treated yourself. The 2014 guideline's own summary of where thinking moved is that degeneration of the cuff tendons now carries more weight, with that degeneration eventually leading to tears in some shoulders, and the 2026 update folds degenerative changes of the rotator cuff into the definition of the condition itself.

How often those changes turn up in people who have no shoulder pain has now been pooled. Yang and colleagues brought together 17 studies, 19 datasets and 5,988 people across eight countries, all scanned with ultrasound or MRI. In the 10 datasets drawn from people with no symptoms, or who had never sought care for a shoulder, 23.6% had a rotator cuff tear (95% confidence interval 14.2 to 32.9). The pooled figures rose across the age bands, and the authors write that prevalence generally increased with age, while noting the pattern was not strictly ordered in the oldest bands where few studies contributed. They also stress how much the studies disagreed with each other, enough that they argue for population-specific figures rather than one headline number, and their discussion makes the point this page needs: structural findings "do not necessarily correspond to symptomatic disease burden" (Yang and colleagues, 2026).

Two things follow for the cause question. Tendon change is partly what time does, so finding some does not mean you broke something. And because those changes are also found in quiet shoulders, a finding on a scan is not automatically the reason this shoulder hurts now. Whether a painful shoulder holds a tear, and what that would change, is on impingement, bursitis or a rotator cuff tear.

Is it my posture?

Less than almost anyone expects, and the argument with its sources is already written out in the posture section of the main shoulder impingement page, so it is not rebuilt here. The short version: people with this shoulder pain do not share one shoulder blade position, and the rounding of the upper back is much the same in people with and without shoulder pain.

Two ideas get the same word, though, and they need keeping apart. A sustained working position under load, with the arm held out or up for hours, is the item the Dutch guideline's prevention chapter actually lists. How you sit on the sofa is what the posture reviews looked at and found wanting. Posture as a subject of its own is on forward head posture, and shoulder blade control on scapular dyskinesis, where we are equally careful not to call a movement pattern the cause.

What causes shoulder impingement to flare up?

This is the more answerable half of the question, and it is usually what people mean when they ask about causes in week three. No study has set out to rank what sets a flare off in this condition, so nobody can hand you a list with numbers on it. What the guideline gives is the principle: activities that are too long, too heavy or too frequent are the ones to pull back from (Lambers Heerspink and colleagues, 2026), and the NHS puts the same idea in two lines, "do not do things that seem to make it worse" and "do not completely stop using your shoulder".

Look at the two or three days before it started and a change is usually sitting there. Height is the commonest one, because reaching above shoulder level asks the most of this shoulder, so a day of ceilings, high shelves or overhead cupboards shows up that evening. Weight held away from the body is next, which is why a bag of shopping carried at arm's length costs more than a heavier one held in close. Then volume and speed, the sets that went up, the repeats that went from eight to fifteen, the drive where the hand stayed on the wheel for four hours. Then sleep, since a few nights on that side will inflame a shoulder that was quietly settling, and that has a page of its own at sleeping with shoulder impingement. This is the pattern we see, not a measured finding, and it is worth ten minutes of thinking back, not three days.

What to cut back on, and how to carry things while a shoulder is sore, is set out on the parent page, so this section stops at what tends to light the fuse. One kind of flare is a different condition, not a worse week: pain that arrives suddenly, is severe out of proportion to anything you did and is hard to bear can come from a calcium deposit in the tendon, which is calcific tendinitis. Sudden, very bad shoulder pain, or an arm you suddenly cannot lift at all, is on the warning list below with a same-day route, and that route applies whatever the explanation turns out to be.

Can shoulder impingement come and go, and can you stop it coming back?

Coming and going is in the nature of this condition. The condition is defined by pain mainly when you lift the arm, the NHS describes shoulder pain that is often worse while you use the arm, and OrthoInfo adds that the pain is typically at its worst at night. A shoulder that is quiet for a fortnight and loud again after a busy weekend is behaving the way this condition behaves. A shoulder that steps down a level and stays there, with weakness or lost movement rather than more pain, is a different pattern and belongs in front of someone.

On stopping it coming back, we will give you what the sources give and no more. The 2026 update's search for prevention studies came back empty, which is the plainest fact on this page. The 2014 guideline reports evidence that regular sport, more than 3 hours a week for at least 10 months of the year, has a preventive effect on the risk of neck and shoulder complaints, which is a broader group of complaints than this one diagnosis, and it recommends changing the work factors listed above early rather than late. The NHS suggests keeping shoulder exercises going for 6 to 8 weeks to stop the pain returning.

What there is not, anywhere we can read, is a proven program for stopping this from happening again, and we are not going to invent one. The nearest thing is a shoulder that has been built back up to what you actually ask of it, which is what the staged program on the main shoulder impingement page is for, including the overhead work later on with exercises such as wall angels and prone scapular retraction. Your physio, or your physical therapist if you are in the United States, will adjust that to the job, sport or lift you want back.

When shoulder pain needs checking rather than explaining

Some shoulders need an assessment more than they need a theory, and the NHS marks the line clearly: see a doctor if the pain is getting worse or has not improved after 2 weeks, or if it is very difficult to move the arm. It also lists what needs urgent help the same day, including sudden or very bad shoulder pain, being unable to move the arm, pain that started after an injury or accident such as a fall, severe pain in both shoulders, and feeling feverish or unwell.

The warning list below carries the exact wording for each of those, sorted by how fast to act, so this is only the map. Straight to emergency services or an emergency department: chest pain or tightness spreading to the arm, neck or jaw, a shoulder that has slipped out of place or looks a different shape, and an arm, hand or fingers that turn cold, pale, blue or gray compared with the other side. The same day: sudden very bad pain or an arm you cannot lift at all, pain that began with a fall or a pull where the arm has been weak since, a hot, red and swollen shoulder with a fever, severe pain in both shoulders that has come on recently, and numbness or pins and needles in the arm that is not going away. Within a few days to a day or two: the morning stiffness pattern in the over-fifties, night pain that nothing eases and that keeps worsening, and new shoulder pain if you have had cancer or are losing weight without meaning to.

Other explanations sit on other pages, and sorting between them is an assessment, not a reading exercise. A shoulder that stops short in every direction however it is moved, including when someone else lifts the arm for you, fits frozen shoulder better than this condition. Pain that only announces itself as the arm goes up is covered from the symptom end on shoulder pain when lifting your arm, and the rest of the shoulder problems we cover are listed under shoulder conditions.

If you keep two sentences from this page, keep these. Nobody can tell you the one cause of your shoulder impingement, because the condition does not work that way and the guidelines say as much. What can usually be found, and changed, is what the shoulder was asked to do in the days before it got loud.

Related exercise programs

• Shoulder impingement

Common questions

References

1. Diercks R, Bron C, Dorrestijn O, et al. Guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review by the Dutch Orthopaedic Association. Acta Orthopaedica. 2014;85(3):314-322. https://doi.org/10.3109/17453674.2014.920991
2. Lambers Heerspink FO, Veen EJD, Dorrestijn O, et al. Update of guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review by the Dutch Orthopedic Association. Part 1: preventive measures, diagnostics, and non-surgical treatment of subacromial pain syndrome. Acta Orthopaedica. 2026;97:91-98. https://doi.org/10.2340/17453674.2026.45365
3. Yang Y, Fu X, Meng S, Zhong M. Prevalence of rotator cuff tears: a systematic review and meta-analysis of observational studies. Frontiers in Physiology. 2026;17:1876217. https://doi.org/10.3389/fphys.2026.1876217
4. NHS. Shoulder pain. Page last reviewed 22 May 2023. https://www.nhs.uk/symptoms/shoulder-pain/
5. American Academy of Orthopaedic Surgeons. Shoulder impingement/rotator cuff tendinitis. OrthoInfo. https://www.orthoinfo.org/en/diseases--conditions/shoulder-impingementrotator-cuff-tendinitis/

Written and checked by the PocketPhysio editorial team. Last updated 2026-10-18.

This page is general education, not a diagnosis or a personal treatment plan. See a physiotherapist or doctor for advice about your own situation.